3. The History of Single Sideband Modulation
Comparison of SSB with AM
Further examination of figure 1 provides an insight into the significance of relative power distribution among the components which make up AM and SSB signals. As previously stated, the desired intelligence is transmitted in the sidebands. Notice, however, that in a conventional AM signal considerably more power is contained in the carrier than is present in either sideband. Furthermore, the remaining power content of the signal is split equally between two sidebands which contain identical information. In contrast, suppression of the carrier and one sideband in an SSB signal allows a concentration of power in the remaining sideband. The result is that SSB makes more effective use of amplifier input capability.
The use of SSB makes other significant economies possible. At higher power levels in particular, plate modulators and power supplies for AM transmitters are relatively expensive. A considerable cost reduction in these areas is possible with SSB, because the high-level modulator and the power supply load imposed by the need to amplify a high-energy carrier are eliminated. Since speech input signals are intermittent in nature, further savings can be made in the cost of power supply components for an SSB transmitter. Consequently, high-power amateur SSB transmitters are less expensive than comparable AM transmitters.
Signal-To-Noise Comparison of SSB and AM
The relative performance of SSB and AM systems can be evaluated by comparing the transmitter power required by each to produce a given signal-to-noise (s/n) ratio at the receiver under ideal propagating conditions. This comparison is useful since, neglecting frequency response, the s/n ratio determines the intelligibility of a received signal.
Figure 1-3A shows the power spectrum for an AM transmitter rated at one unit of carrier power. With 100-percent sine-wave modulation, such a transmitter produces 1.5 units of RF power. The additional 0.5 unit of power is furnished by the modulator and is distributed equally between the two sidebands. This AM transmitter is compared with an SSB transmitter rated at 0.5 unit of peak-envelope power (PEP). Peak-envelope power is defined as the RMS power developed at the crest of the modulation envelope.
When the RF signal is demodulated in the AM receiver an audio voltage develops which is equivalent to the sum of the upper- and lower-sideband voltages, in this case 1 unit of voltage. This voltage represents the output from a diode detector as normally used for AM reception. Such detection is called coherent detection because the voltages of the two sidebands are added in the detector. When the RF signal is demodulated in the SSB receiver, an audio voltage of 0.7 unit develops which is equivalent to the transmitted upper-sideband signal. This signal normally is demodulated by re-inserting the carrier at the detector. In a superheterodyne receiver, the received signal is translated to a new frequency before audio detection takes place. Therefore the re-inserted carrier must be at a frequency which maintains approximately the same frequency relationship to the sideband as in the transmitted signal. If a broadband noise level is chosen as 0.1 unit of voltage per 6 kc bandwidth, the AM bandwidth, the same noise level is equal to 0.07 unit of voltage per 3 kc bandwidth, the SSB bandwidth. These values represent the same noise power level per kc of bandwidth, that is, 0.12 divided by 6 is equal to 0.072 divided by 3. The s/n ratio for the AM system is 20 log s/n in terms of voltage, or 20 dB. For the SSB system the s/n ratio is also 20 dB. Therefore the 0.5 power unit of rated PEP for the SSB transmitter produces the same signal intelligibility as the 1 power unit of rated carrier power for the AM transmitter .
In summary it can be stated that, under ideal propagating conditions but in the presence of broadband noise, an SSB signal and an AM signal provide equal s/n ratios at the receiver if the total sideband power contained in each of the signals is equal. This means that, to perform under these conditions as well as an SSB transmitter of given PEP rating, an AM transmitter requires twice that figure in carrier power rating.
Antenna Voltage Comparison of SSB and AM
The use of multiband antennas is rather common on the amateur bands. Frequently these antennas employ resonant tuned circuits or traps, to disconnect portions of the antenna on certain bands, thus providing multiple resonant frequencies. Several types of triband beam antennas are typical examples. In operation these traps are subjected to high voltages which are functions of antenna input power. The maximum voltage rating of the traps places a limitation on the amount of peak power which may be fed into the antenna. As an example, if the peak power rating of a given antenna is 1000 watts, an AM transmitter used with this antenna must be limited to 250 watts carrier output power. This is true because the PEP of an AM signal, at 100-percent modulation, is four times the carrier power. In comparison, an SSB transmitter rated at 1000 watts PEP output, all of which is sideband power, may be used with this same antenna.
An interesting sidelight related to this comparison involves television and broadcast interference. For a given amount of total sideband power output, the PEP output of a conventional AM transmitter must be eight times that of an equivalent SSB transmitter. Therefore, even though such an AM transmitter is no more effective in transmitting the desired intelligence, it is more likely to cause fundamental overload interference than is the SSB transmitter.
Advantage of SSB with Selective Fading
The signal-to-noise comparison between SSB and AM, as discussed in a previous paragraph, is based upon ideal propagating conditions. Over many transmission paths, however, signals are subject to a phenomenon known as selective fading. This type of fading is caused by multipath propagation and is characterized by selective attenuation of the individual components which make up the transmitted signal. An AM signal is subject to severe distortion under these conditions principally because of its dependence upon a received carrier.
A few of the multiple paths over which a transmitted signal could be propagated are illustrated in figure 3. On the lower h-f bands, the signal often reaches the receiver by means of both sky waves and ground waves. Multiple sky-wave paths predominate on the upper h-f bands. As indicated in this illustration, the propagating medium is not a single, uniform reflecting surface. It is subject to continual variations in density, stratification, and refractive index. The effect of these propagating conditions is to cause varying instantaneous phase relationships among identical signal components arriving at the receiver over different paths. At a given moment, some of the arriving signal components are out of phase with those propagated over another path. The net result is that nulls are created in the spectrum of the received signal, and these nulls move across the signal spectrum as the propagating medium changes. The effect is somewhat like tuning a narrow rejection notch across the receiver passband. Some loss among the sideband components can be tolerated, since the practical effect is to upset only the amplitude and frequency response of the received signal. With a conventional AM signal, attenuation of the carrier by these selective nulls causes a serious problem. In this situation the carrier voltage at the receiver can be appreciably less than the sum of the two sideband voltages. Consequently the RF envelope does not retain its transmitted shape, and distortions severe upon demodulation. Frequently the fading is sufficient to render the signal unintelligible a large percentage of the time even though average signal strength may be quite high. Distortion caused by selective fading can be largely avoided by the use of the exalted carrier technique, but this method requires the local carrier to be phase locked to the transmitted carrier. The means to accomplish this, such as automatic-frequency control of the receiver BFO, requires a substantial amount of added circuitry. A SSB signal is not degraded significantly by selective fading. Since the received signal is not dependent upon a received carrier, no degradation results from the loss of carrier power. Transmission of only one sideband removes the necessity for precise carrier re-insertion at the receiver. Selective fading within the one sideband of a SSB signal alters only the amplitude and frequency response of the signal. Rarely does it cause enough distortion to make the signal unintelligible.
Go to the Next
Back to the
© Copyright by DJ4BR, Ing. Peter Weber, Am Eisenbrand 18, D-40667 Meerbusch, Germany